Combined antiretroviral therapy (cART) has dramatically changed the HIV epidemic, delaying disease and prolonging life. With increasing emphasis on development of strategies to eradicate HIV from latent reservoirs including the CNS, it is critical to evaluate therapeutic approaches in established animal models of HIV latency. To do so, we developed and characterized a SIV model of cART that reduced viral load in peripheral blood and cerebrospinal fluid to undetectable levels. The value of this model is : 1) CD4+ T cells and monocyte/ macrophages are infected, 2) tissues including the brain harbor latent viral DNA, and 3) the number of latently infected resting CD4+ cells in the blood and lymphoid tissues is comparable to that in HIV-infected patients on cART. While SIV RNA in brain was dramatically reduced by cART, SIV DNA levels in brain were unchanged compared to untreated SIV-infected macaques and inflammatory markers remained elevated. In sum, our studies illustrate that cART that suppresses CSF and plasma viral load does not target the CNS latent DNA reservoir. CCR5 inhibitors are promising anti-HIV drug candidates with potential beneficial CNS effects. The CCR5 inhibitor maraviroc (MVC) has high CNS penetrance (CPE = 1.0) and minimal neurotoxicity in comparison with other classes of antiretrovirals. As R5-tropic HIV predominates in the CNS and intermittent HIV replication may persist in the brain despite cART, MVC treatment could block infection of additional cells in the CNS. MVC also may dampen immune activation of resident effector cells in the brain, including microglia and astrocytes, and decrease recruitment of leukocytes to the CNS. Provocative SIV studies by our group evaluated the impact of CCR5 inhibition on SIV-induced CNS damage. In SIV-infected rhesus macaques, maraviroc monotherapy significantly reduced CNS SIV DNA levels and lowered key CNS inflammatory responses in sharp contrast with animals treated with cART. Because of these findings, this proposal focuses on intensification of cART therapy in SIV-infected pigtailed macaques by adding maraviroc with the goal of virus eradication. Our hypothesis is that adding the CNS penetrant CCR5 inhibitor maraviroc to cART in SIV- infected macaques will be much more effective than cART alone in A) reducing viral DNA reservoirs in brain and B) delaying virus reactivation upon withdrawal of cART. These effects are attributable to MVC's ability to both block infection of new cellular targets in the CNS and inhibit pro-inflammatory signaling through CCR5. Aim 1 is to determine whether adding MVC to cART reduces viral DNA and impairs reactivation of latent virus in macrophages/ microglia and astrocytes in the CNS and resting CD4+ cells and monocyte/macrophages in peripheral tissues and blood of SIV-infected macaques. Aim 2 will compare the ability of cART+MVC versus cART alone regimens to prevent or delay reactivation of virus from the CNS reservoir in SIV-infected macaques after cessation of therapy. Aim 3 is to determine whether continuing MVC therapy after stopping cART suppresses reactivation of virus from the CNS reservoir or peripheral reservoirs.